Left ventricle epicardium estimation in medical diagnostic imaging
The left ventricle epicardium is estimated in medical diagnostic imaging. C-arm x-ray data is used to detect an endocardium at different phases. The detected endocardium at the different phases is compared to sample endocardiums at different phases. The sample endocardiums have corresponding sample epicardiums. The transformation between the most similar sample endocardium or endocardiums over time and the detected endocardium over time is applied to the corresponding sample epicardium or epicardiums. The transformed sample epicardium over time is the estimated epicardium over time for the C-arm x-ray data.
1. A method for left ventricle epicardium estimation in medical diagnostic imaging, the method comprising:
acquiring C-arm x-ray data representing a left ventricle of a patient over time;
detecting an endocardium over time from the C-arm x-ray data;
identifying, with a processor, at least one motion template from a library of motion templates, the identifying being based on a similarity of the endocardium over time detected from the C-arm x-ray data with an endocardium of the at least one motion template, wherein the similarity is calculated by calculating a difference between the endocardium detected from the C-arm x-ray data and the endocardium of each of the motion templates of the library at each of a plurality of heart cycle phases, averaging the differences over the heart cycle phases for each motion template, and identifying the at least one motion template as having a least of the average difference;
calculating, with the processor, a deformation field as a function of a difference between the endocardium of the at least one motion template and the endocardium detected from the C-arm x-ray data;
applying, with the processor, the deformation field to an epicardium represented by the at least one motion template; and
displaying an image of the left ventricle of the patient, the image being a function of the epicardium as deformed by the applied deformation field.
2. The method of claim 1 wherein acquiring comprises acquiring with a single sweep of a C-arm.
3. The method of claim 1 wherein identifying comprises identifying from the library, the library of motion templates generated from computed tomography or magnetic resonance data.
4. The method of claim 1 wherein identifying comprises identifying from the library comprising at least one hundred motion templates, each of the motion templates representing different patients.
5. A method for left ventricle epicardium estimation in medical diagnostic imaging, the method comprising:
acquiring C-arm x-ray data representing a left ventricle of a patient over time;
detecting an endocardium over time from the C-arm x-ray data;
identifying, with a processor, at least five motion templates from a library of motion templates, the identifying being based on a similarity of the endocardium over time detected from the C-arm x-ray data with an endocardium of the at least five motion templates, wherein the similarity is calculated between the detected endocardium and the library of motion templates;
calculating, with the processor, a deformation field as a function of differences, including the difference, between the endocardium of the at least five motion templates and the endocardium detected with the C-arm x-ray data;
applying, with the processor, the deformation field to an epicardium represented by the at least five motion templates, wherein applying further comprises estimating, with a weighted average function, the epicardium of the C-arm x-ray data from the epicardium of the at least five motion templates as deformed by the deformation field; and
displaying an image of the left ventricle of the patient, the image being a function of the epicardium as deformed by the applied deformation field.
6. The method of claim 1 wherein calculating comprises determining the difference between the endocardiums with a non-rigid deformation.
7. The method of claim 1 wherein calculating comprises interpolating with a thin plate spline function.
8. The method of claim 1 wherein the at least one motion template comprises a first mesh as the endocardium and a second mesh as the epicardium, and wherein applying comprises estimating the epicardium of the C-arm x-ray data from the epicardium of the at least one motion template as deformed by the deformation field.
9. In a non-transitory computer readable storage medium having stored therein data representing instructions executable by a programmed processor for left ventricle epicardium estimation in medical diagnostic imaging, the storage medium comprising instructions for:
selecting a first endocardium surface from a library of endocardium surfaces based on matches in shape over time from a second endocardium surface for a patient, by calculating a difference between the second endocardium surface for a patient and the endocardium of each of the motion templates of the library of endocardium surfaces at each of a plurality of heart cycle phases, averaging the differences over the heart cycle phases for each motion template, and selecting the at least one motion template as having a least of the average difference;
determining a distortion of the selected first endocardium surface to the second endocardium surface;
applying the distortion to a first epicardium surface associated with the selected endocardium surface and from the library; and
estimating a second epicardium surface of the patient as the first epicardium surface.
10. The non-transitory computer readable storage medium of claim 9 wherein selecting comprises:
determining a level of match between the second endocardium surface of the patient and each of a plurality of endocardium surfaces in the library;
repeating the determining for each of a plurality of heart phases, the endocardium surfaces of the library and the second endocardium surface being represented over the heart phases;
calculating a spatial-temporal match for each of the different endocardium surfaces of the library from the second endocardium surface as a function of the levels from the heart phases;
selecting the first endocardium surface from the library as a function of the spatial-temporal match.
11. The non-transitory computer readable storage medium of claim 9 wherein selecting comprises selecting the first endocardium surface and a plurality of third endocardium surfaces from the library, wherein determining the distortion comprises determining third distortions of the selected third endocardium surfaces, and wherein applying comprises weighted averaging based on the distortion and the third distortions.
12. The non-transitory computer readable storage medium of claim 9 wherein the second endocardium surface is detected from C-arm x-ray data obtained in a single sweep, wherein the endocardium surfaces of the library are generated from computed tomography or magnetic resonance data.
13. The non-transitory computer readable storage medium of claim 9 wherein determining the distortion comprises transforming the first endocardium surface towards the second endocardium surface with a non-rigid transform.
14. The non-transitory computer readable storage medium of claim 9 wherein estimating comprises using the first epicardium surface as the second epicardium surface.
15. A system for left ventricle epicardium estimation in medical diagnostic imaging, the system comprising:
a memory configured to store C-arm x-ray data representing a left ventricle over at least one heart cycle; and
a processor configured to determine endocardium motion based on the C-arm x-ray data and a template of endocardium motion based on a similarity between the C-arm x-ray data endocardium motion and the template of endocardium motion, wherein the processor selects the template from a library of templates, the selection based on a similarity of the template to an endocardium detected from the C-arm x-ray data, to determine the endocardium motion based on endocardium shapes, over a plurality of phases of the at least one heart cycle, for the template and from the C-arm x-ray data, and configured to estimate an epicardium location over the at least one heart cycle from the endocardium motion; and wherein the similarity is calculated by calculating a difference between the endocardium detected from the C-arm x-ray data and the endocardium of each of the motion templates of the library at each of a plurality of heart cycle phases, averaging the differences over the heart cycle phases for each motion template, and identifying the at least one motion template as having a least of the average difference;
the processor is configured to calculate a deformation field as a function of a difference between the endocardium of the at least one motion template and the endocardium detected from the C-arm x-ray data; wherein
the processor is configured to apply the deformation field to an epicardium represented by the at least one motion template; and
a display configured to display an image of the left ventricle of the patient, the image being a function of the epicardium as deformed by the applied deformation field.
16. The system of claim 15 wherein the processor is configured to estimate the epicardium location based on a transform of the template of endocardium motion to the endocardium motion based on the C-arm x-ray data.
17. The system of claim 15 wherein the processor is configured to estimate the epicardium location from a plurality of templates of endocardium motion.